Download Cardiac magnetic resonance imaging-based myocardial strain study

non-invasive imaging
ORIGINAL ARTICLE
Cardiology Journal
2016, Vol. 23, No. 3, 270–280
DOI: 10.5603/CJ.a2016.0023
Copyright © 2016 Via Medica
ISSN 1897–5593
Cardiac magnetic resonance imaging-based
myocardial strain study for evaluation
of cardiotoxicity in breast cancer patients
treated with trastuzumab: A pilot study
to evaluate the feasibility of the method
Shintaro Nakano1, Masahiro Takahashi2, Fumiko Kimura2, Taiki Senoo 2, Toshiaki Saeki3,
Shigeto Ueda3, Jun Tanno1, Takaaki Senbonmatsu1, Takatoshi Kasai4, Shigeyuki Nishimura1
1
Department of Cardiology, International Medical Center, Saitama Medical University, Saitama, Japan
2
Department of Diagnostic Radiology, International Medical Center,
Saitama Medical University, Saitama, Japan
3
Department of Breast Oncology, International Medical Center,
Saitama Medical University, Saitama, Japan
4
Cardiovascular Respiratory Sleep Medicine, Department of Cardiovascular Medicine,
Juntendo University Graduate School of Medicine, Tokyo, Japan
Abstract
Background: Trastuzumab, used to treat breast cancer overexpressing human epidermal
growth factor receptor 2, may be cardiotoxic. Cardiac magnetic resonance (CMR) imaging
with myocardial strain studies has been used to evaluate subclinical biventricular myocardial
changes, however, its clinical utility during chemotherapy has not been evaluated.
Methods: The clinical outcomes, CMR and cardiac biomarkers of 9 women aged 62.3 ± 12.6
years with early or locally advanced breast cancer were evaluated at baseline, and at 3, 6 and
12 months after the initiation of trastuzumab.
Results: None of the patients developed heart failure or elevated serum cardiac biomarkers.
Global left ventricular (LV) peak systolic longitudinal and circumferential strains were significantly decreased at 6 months (longitudinal strains, –21.1 ± 1.7% [baseline] vs. –19.5 ± 1.0%
[6 months], p = 0.039, and circumferential strains, –23.4 ± 1.8% [baseline] vs. –21.6 ± 2.5%
[6 months], p = 0.036). These changes were analogous to those observed in the LV ejection
fraction. Right ventricular (RV) free wall peak systolic circumferential strains were decreased
at 6 months (–20.9% ± 2.4% [baseline] vs. –19.1% ± 2.3% [6 months], p = 0.049), whereas
RV longitudinal strains and ejection fraction remained unchanged. The LV longitudinal strain
was the most reproducible of the 4 peak strain parameters.
Conclusions: The LV longitudinal and circumferential strains measured by CMR decreased
during trastuzumab therapy, although their predictive value for later heart failure or association with RV parameters was not determined. These techniques may be a useful means of diagnosing and monitoring trastuzumab-related cardiotoxicity. (Cardiol J 2016; 23, 3: 270–280)
Key words: cardiac magnetic resonance, myocardial strain, trastuzumab,
breast cancer, cardiotoxicity
Address for correspondence: Shintaro Nakano, MD, Department of Cardiology, International Medical Center, Saitama
Medical University 1397-1 Yamane, Hidaka, Saitama 350-1298, Japan, tel: +81-42-984-4111, fax: +81-42-984-4591,
e-mail: [email protected]
Received: 24.12.2015
270
Accepted: 07.05.2016
www.cardiologyjournal.org
Shintaro Nakano et al., CMR for trastuzumab-related cardiotoxicity
Introduction
Trastuzumab, a humanized monoclonal antibody against the extracellular domain of human
epidermal growth factor receptor 2 (HER2), improves clinical outcomes in patients with breast
cancer overexpressing HER2. Adjuvant therapy
using trastuzumab reportedly reduces the risk
of relapse by about a half and the risk of death by
a third; a substantial body of evidence for its benefits has led to trastuzumab becoming a first line
treatment for early-stage HER2-positive breast
cancer [1, 2].
Although generally well tolerated, trastuzumab is cardiotoxic in 2–27% of patients; the
incidence varies according to the definition used
and concomitant chemotherapy [1, 3]. The Cardiac
Review and Evaluation Committee (CREC) that
examined trastuzumab-associated cardiotoxicity defined cardiotoxicity as a ≥ 5% reduction
in left ventricular (LV) ejection fraction (LVEF)
to < 55% in symptomatic patients, or a ≥ 10%
reduction in LVEF to < 55% in asymptomatic
patients [4]. Nevertheless, LVEF measured by
echocardiography may not be sufficiently specific
or sensitive to detect cardiotoxicity reliably [5, 6].
Deformation parameters such as myocardial strain
have been used to detect subclinical myocardial
dysfunction before reduced LVEF is identifiable
on echocardiography in patients with breast cancer
receiving trastuzumab [6–9].
Cardiac magnetic resonance (CMR) imaging
provides excellent images of the heart, allowing
3-dimensional assessments of LVEF and ventricular volumetric function [10]. Although its high
cost and lack of availability precludes its routine
clinical use, CMR has been validated for serial
monitoring of LVEF and volume in patients with
breast cancer receiving trastuzumab, as well as for
the characterization of myocardial tissue changes
[7, 11, 12]. Recently, CMR has been also used to monitor right ventricular (RV) ejection fraction (RVEF)
in cardiotoxicity [13]. Additionally, CMR can now
be used to measure deformation parameters in
conjugation with strain-encoded imaging (SENC)
[14], in which objective color-coded biventricular
longitudinal and circumferential myocardial strain
patterns are obtained in several heartbeats and
breath-holds.
We hypothesized that CMR-based measurement of myocardial strains, in addition to the
conventional biventricular CMR parameters, may
be a reliable means of assessing change in cardiac
function during chemotherapy. Our pilot study
aimed to establish whether CMR with myocardial
strain studies might have clinical utility in patients
with breast cancer treated with trastuzumab.
Methods
Patient selection and study design
Of patients newly diagnosed with HER2-overexpressing breast cancer between April 2012 and
October 2014 in our institute, 16 women consented
to participate in our pilot study. The exclusion
criteria were: 1) artificial pacemaker implantation,
2) chronic arrhythmia such as atrial fibrillation,
3) chronic kidney disease (estimated glomerular filtration rate < 30 mL/min/1.73 m 2), and
4) metastatic cancer. Intravenous trastuzumab was
administered at a loading dose of 8 mg/kg followed
by 6 mg/kg every 3 weeks (a total of 18 doses) in all
patients. Physical examination, echocardiography,
CMR and blood tests were performed before the initiation of trastuzumab, and 3 months (± 6 weeks),
6 months (± 6 weeks) and 12 months (± 8 weeks)
afterwards. Cardiac function was evaluated by CMR
at each of the 4 time points, and the relationships
with echocardiographic parameters, biomarkers
and clinical indices were evaluated.
Data collected at baseline
Patients’ demographic characteristics, including age and body mass index, were recorded at
baseline. The cardiovascular history was taken
and risk factors, including hypertension, dyslipidemia, diabetes mellitus, chronic atrial fibrillation,
chronic kidney disease, coronary artery disease
(myocardial infarction or angina pectoris) and current smoking were assessed (see Supplementary
Materials). Cancer-related variables, including
side of the breast cancer, stage (early or locally
advanced), surgical treatment, radiotherapy, and
other concomitant chemotherapy were recorded.
Vital signs, including systolic and diastolic blood
pressure, and heart rate at rest were recorded.
Echocardiographic variables, including LV enddiastolic dimension, end-systolic dimension, and
intra-ventricular septum and LV posterior wall
thickness were measured, and the presence of
significant valvular disease was evaluated. LVEF
was calculated using the Teichholz and modified
Simpson’s methods [15, 16].
Cardiac magnetic resonance imaging
Cardiac magnetic resonance imaging was
performed within 28 days of echocardiography
using a 3.0 Tesla MR imaging scanner (Archieva
www.cardiologyjournal.org
271
Cardiology Journal 2016, Vol. 23, No. 3
Figure 1. Measurement of peak systolic strain. Myocardial strain was measured using strain-encoded cardiac magnetic
resonance imaging (SENC-CMR). The short-axis view shows longitudinal myocardial strain and the 4-chamber view
shows circumferential strain. Areas of myocardium with severely impaired or absent strain are shown in white, and
areas of myocardium with normal strain are shown in red. Global strain was defined as the mean of peak systolic strain
of all allocated segments; A–C. Longitudinal strain, in the basal, middle and apical levels; D. Circumferential strain.
3.0T TX, Philips Medical Systems, Best, The
Netherlands) with a 32-element phased-array
cardiac coil. Cine studies using steady-state free
precession sequence in the LV vertical long-axis,
4-chamber view, and sequential short-axis slices
from the mitral valve to the apical levels were obtained at 4 time points (Supplementary Table S1).
T2-weighted short tau inversion recovery (STIR)
and late-gadolinium enhancement (LGE) images
after intravenous administration of 0.15 mmol/kg
meglumine gadoterate (Magnescope; Guerbet
Japan, Tokyo, Japan) were obtained at baseline only, to exclude pre-existing cardiac disease or irreversible myocardial injury that had
developed after previous chemotherapy. LV
end-diastolic volume index (LVEDVI in mL/m2
indexed by body surface area), LV end-systolic
volume index (LVESVI in mL/m2) and LV mass
index (LVMI in g/m2) were measured as previously
described [17] using a Philips Extended MR Workspace 2.6.3.4 (Philips Medical Systems). Briefly,
using the stack of short-axis cine images, the most
basal and apical slices of the LV were determined,
and LV endocardial and epicardial borders were
manually traced to obtain LVEDVI, LVESVI and
LVMI. LVEF was calculated using the blood volume
method, which excluded portions of papillary and
trabecular muscle. Using the same images, RV
end-diastolic volume index (RVEDVI in mL/m2) and
272
RV end-systolic volume index (RVESVI in mL/m2)
were measured [18]; RVEF was calculated using
the blood volume method.
Myocardial strains were evaluated using SENCCMR (version 3.0; Diagnosoft, Palo Alto, CA, Fig. 1),
based on the acquisition of two images of differing
frequency modulation; superposition of low- and
high-tuning images allowed a SENC strain map to
be plotted, which was overlaid pixel-by-pixel on the
anatomical images [14, 19]. In contrast with conventional tagging, in SENC-CMR the gradient is applied
in the slice selection direction [14], therefore, the
short-axis view shows longitudinal myocardial strain
and the 4-chamber view shows circumferential
strain. Areas of myocardium with severely impaired
or absent strain are shown in white, and areas of
myocardium with normal strain are shown in red.
LV longitudinal strain was measured at the basal,
middle, and apical levels of the short-axis view;
a region of interest was plotted in the myocardial
middle-layer of each myocardial segment and the
minimum value during systole (i.e., peak systolic
strain) was recorded. Using the same images, RV
longitudinal and circumferential strains were measured in regions of interest plotted in the RV free wall
using a method modified from a previous report [20].
Global strain was defined as the mean peak systolic
strain of all allocated segments (16 segments for
LV longitudinal strain, 6 for circumferential strain,
www.cardiologyjournal.org
Shintaro Nakano et al., CMR for trastuzumab-related cardiotoxicity
7 segments for RV free wall longitudinal strain, and
3 for circumferential strain). Left ventricular wall
motion at baseline was evaluated by the agreement
of two observers using a 17-segment model [21].
The detailed CMR protocol is described in Supplementary Table S1.
Other data collected
Clinical signs and symptoms of heart failure
(such as dyspnea and peripheral edema), and the
serum concentrations of troponin-I (Architect
STAT Troponin-I assay, Abbott Diagnostics,
Illinois, USA), ventricular myosin light chain (SRL,
Tokyo, Japan), and plasma B-type natriuretic peptide (BNP) (TOSOH, Osaka, Japan) were measured
at the same intervals (see Supplementary Materials). Values below the lower limit of detection
were treated as zero. LVEF was also measured by
echocardiography at each time point.
Statistical analysis
Continuous data are expressed as the mean
± standard deviation (SD) and categorical data as
the number (proportion). Independent continuous
variables at the 3 different time points (3, 6 and
12 months) were compared with the baseline data using
the paired t-test. Intra- and inter-observer variability
was evaluated for reproducibility using Pearson’s
correlation analysis; standardized dispersion was
expressed as coefficient of variation. Inter-modality
comparison between CMR and echocardiography
was also evaluated using Pearson’s correlation
analysis. A p value < 0.05 was considered statistically significant. All analyses were performed using
JMP V.10.0.0 (SAS Institute Inc., Cary, NC).
Ethics approval
The study was conducted according to the
Declaration of Helsinki and approved by the Institutional Review Board of the International Medical
Centre of Saitama Medical University, Japan (reference number 11-067).
Results
Baseline characteristics, and changes
in echocardiographic findings and cardiac
biomarkers from baseline
Of the 16 women enrolled, 7 were excluded as they did not undergo CMR at each of the
4 time points due to: progression of cancer (n = 2);
interruption by surgery (n = 3), and withdrawal
of consent to participate (n = 2). The data of remaining 9 patients were analyzed. The mean age
of the patients was 62.3 ± 12.6 years (Table 1).
Two had essential hypertension (1 was treated
with an angiotensin converting enzyme inhibitor),
1 had diabetes mellitus and 1 had dyslipidemia,
but none had previous or co-existing heart failure.
Eight (88.9%) patients had a right-sided tumor,
and 8 (88.9%) underwent either partial or total
mastectomy. Intravenous epirubicin was administered prior to trastuzumab in 6 (66.7%) patients,
with a cumulative dose of 360 mg/m2 in 5 patients
and 342 mg/m2 in 1 patient (in whom the dose was
reduced owing to inflammation of the throat). Echocardiography at baseline demonstrated preserved
LVEF (71.8% ± 7.4%, range: 62.7–86.2%) and
the absence of significant valvular disease in all
patients. The serum concentrations of troponin I,
myosin light chain, and plasma BNP were not
elevated at baseline.
Throughout the observation period, none of
the patients developed symptoms of heart failure,
nor did any exhibit an LVEF < 55% measured by
echocardiography (Supplementary Table S2). In all
patients, the serum concentrations of troponin I
and myosin light chain remained within their normal ranges, and the plasma concentration of BNP
remained lower than £ 100 pg/mL, the validated
cutoff value for suspected acute heart failure [22].
In 1 patient, borderline LVEF reduction was treated
with an oral b-blocker, but this was discontinued
almost immediately because of faintness.
Change in CMR parameters
There was no evidence of myocardial LGE
or high signal intensity in T2-weighted short tau
inversion recovery images, or abnormal LV motion
at baseline in any of the participants. The mean
LVESVI increased 3 months after the initiation of
trastuzumab and remained high until 12 months.
The LVEF (68.4 ± 6.6%, range 57.3–80.8% at baseline) significantly decreased by 6 months (61.7 ±
± 8.7%, p = 0.004 compared with baseline), and
remained low at 12 months (62.9 ± 7.8%, p = 0.002
compared with baseline). Over the 12-month study
period, minimum LVEF measured by CMR fell to
54.3% and 53.9% in 2 cases, but the corresponding
LVEF measured by echocardiography was 61.9% and
55.9%, respectively. Other ventricular parameters
such as the LVEDVI and LVMI remained unchanged
throughout the observation period. The conventional RV parameters, such as RVEDVI, RVESVI,
and RVEF remained unchanged (Fig. 2).
Global LV peak systolic longitudinal and circumferential strains significantly decreased by
6 months (longitudinal strain, –21.1 ± 1.7% at
www.cardiologyjournal.org
273
Cardiology Journal 2016, Vol. 23, No. 3
Table 1. Baseline characteristics (total: n = 9).
Demographics
Age [years]
62.3 ± 12.6
Body mass index [kg/m ]
2
22.1 ± 3.4
Cardiovascular history and risk factors
Hypertension
2 (22.2%)
Diabetes mellitus
1 (11.1%)
Dyslipidemia
1 (11.1%)
Chronic atrial fibrillation
0 (0.0%)
Chronic kidney disease
0 (0.0%)
Coronary artery disease
0 (0.0%)
Current smoking
0 (0.0%)
Cancer-related variables
Breast cancer side:
Right
Bilateral
Stage:
8 (88.9%)
0 (0.0%)
0 (0.0%)
Early
3 (37.5%)
Locally advanced
5 (62.5%)
Metastatic
Partial mastectomy + Sn
3 (33.3%)
Partial mastectomy + Ax
0 (0.0%)
Total mastectomy + Sn
2 (22.2%)
Total mastectomy + Ax
3 (33.3%)
5 (55.6%)
Other chemotherapy:
Epirubicin
6 (66.7%)
Docetaxel
3 (33.3%)
Cyclophosphamide
8 (88.9%)
Carboplatin
1 (11.1%)
Vital signs at baseline
Systolic BP [mm Hg]
125.6 ± 9.2
Diastolic BP [mm Hg]
79.4 ± 9.2
Heart rate [bpm]
77.8 ± 7.9
Echocardiographic parameters at baseline
LV end-diastolic dimension [mm]
43.2 ± 2.9
LV end-systolic dimension [mm]
25.3 ± 2.7
LV ejection fraction [%]
71.8 ± 7.4
Intra-ventricular septum thickness [mm]
9.5 ± 1.1
LV posterior wall thickness [mm]
9.3 ± 0.9
Significant valvular disease
0 (0.0%)
Cardiac biomarkers at baseline
Troponin-I [ng/mL]
0.02 ± 0.00
Myosin light chain [ng/mL]
1.33 ± 0.41
B-type natriuretic peptide [pg/mL]
23.1 ± 13.5
Continuous variables are described as the mean ± standard deviation and categorical variables as number (proportion); Ax — axillary
clearance; BP — blood pressure; LV — left ventricular; Sn — sentinel
lymph node biopsy
274
Discussion
0 (0.0%)
Surgery:
Radiotherapy
baseline vs. –19.5 ± 1.0% at 6 months, p = 0.039;
circumferential strains, –23.4 ± 1.8% at baseline
vs. –21.6 ± 2.5% at 6 months, p = 0.036). RV
free wall peak systolic circumferential strains had
decreased by 6 months (–20.9 ± 2.4% at baseline
vs. –19.1 ± 2.3% at 6 months, p = 0.049), while
longitudinal strain remained unchanged (Fig. 2;
Supplementary Table S3). Images demonstrating
a typical reduction in LV longitudinal strain are
shown in Figure 3.
The intra-observer (measured by Dr. Nakano)
and inter-observer (measured by Drs. Nakano and
Tanno) correlation coefficients, and the coefficient
of variation are shown in Table 2. The correlation
coefficients between the LVEF measured by CMR
and that by echocardiography using the modified
Simpson method (26 of 36 studies), and CMR and
echocardiography using the Teichholz method were
0.58 (p = 0.004) and 0.46 (p = 0.004), respectively
(Supplementary Fig. S1).
Although the number of patients was limited,
our pilot study is the first to have used CMR-based
myocardial strain studies to undertake serial evaluations of cardiac function in a cohort of patients
with breast cancer treated with trastuzumab.
We found that global LV peak systolic longitudinal and circumferential strains had significantly decreased by 6 months before recovering by
12 months. These changes were analogous to those
of LVEF measured by CMR. LV longitudinal strain
was most reproducible among the 4 peak strain
parameters. We also found that RV free wall longitudinal strain remained unchanged, whereas
circumferential strain decreased 6 months after
the initiation of trastuzumab. The RV circumferential strain was, however, the least reproducible
and the coefficient of variation was the greatest of
the 4 ventricular peak systolic strain parameters.
Role of conventional CMR in the assessment
of chemotherapy-related cardiotoxicity
Cardiac magnetic resonance imaging has
recently been used to characterize myocardial tissue by LGE for patients with breast cancer being
treated with chemotherapy [7, 23], as the presence of LGE may reflect irreversible myocardial
injury [24]. In our study, no patient showed LGE at
baseline, suggesting that no advanced myocardial
damage had been induced by prior exposure to
epirubicin. The mechanism of trastuzumab-related
cardiotoxicity is unclear; it may be attributable to
www.cardiologyjournal.org
Shintaro Nakano et al., CMR for trastuzumab-related cardiotoxicity
Figure 2. Cardiac magnetic resonance (CMR) imaging parameters. Variables at the three different time points (3, 6 and
12 months) were compared with the baseline data using the paired t-test. Peak systolic myocardial strains are expressed as absolute values; *p < 0.05 vs. baseline; LVEDVI — left ventricular end-diastolic volume index; LVESVI —
left ventricular end-systolic volume index; LVMI — left ventricular mass index; LV-LS — left ventricular longitudinal strain; LV-CS — left ventricular circumferential strain; LVEF — left ventricular ejection fraction; RVEF — right
ventricular ejection fraction; RVEDVI — right ventricular end-diastolic volume index; RVESVI — right ventricular
end-systolic volume index; RV-LS — right ventricular longitudinal strain; RV-CS — right ventricular circumferential strain;
BSL — baseline; M — months.
blockade of the HER2 in cardiomyocytes, representing largely reversible type II cardiotoxicity
without ultrastructural abnormalities [2, 25]. None
of the participants in our study exhibited progressive cardiac dysfunction. Still, the technique may be
a useful means of evaluating irreversible myocardial injury in patients receiving trastuzumab following anthracycline, as prior anthracycline exposure
is reportedly a risk factor for trastuzumab-related
cardiotoxicity alongside concomitant radiation
therapy and general cardiovascular risk factors
(such as older age, obesity, smoking, dyslipidemia,
hypertension, diabetes mellitus and history of
coronary artery disease) [3, 26, 27].
In our cohort of patients with breast cancer,
the LVEF measured by CMR tended to decrease
at 3 months, and had significantly decreased by
6 months after the initiation of trastuzumab. This
finding is consistent with previous studies, which
found the reduction in LVEF was first detectable
at approximately 3 months and reached its peak
between 6 and 9 months after the initiation of trastuzumab [7]. Although there is presently a limited
role for the routine use of CMR in the evaluation
of trastuzumab-related cardiotoxicity, it has several
advantages over echocardiography, such as higher
reproducibility and sensitivity [28]. Indeed, a substantial proportion of our echocardiographic studies
(10 of 36, 27.8%) were of insufficient visual quality
for data acquisition using the modified Simpson’s
method; therefore, only the Teichholz method
was available in all studies. Moreover, the ability
to obtain high-quality, 3-dimensional images with
CMR may be particularly advantageous with breast
cancer patients with left-sided disease, because
the mass effect of cancer or scar tissue wound may
further obscure echocardiographic views or a painful surgical wound may interfere with the proper
placement of the ultrasound probe. Our study
demonstrated a modest intra-modality discrepancy
between LVEF measured by echocardiography and
that by CMR, which was consistent with previous
reports [12, 29, 30]. This inherent intra-modality
discrepancy may be important in patients with borderline cardiotoxicity. In our study, neither of the
2 patients with minimum LVEF < 55% measured
by CMR (54.3% and 53.9% each) would have met
the CREC criteria for cardiotoxicity if only echocar-
www.cardiologyjournal.org
275
Cardiology Journal 2016, Vol. 23, No. 3
Figure 3. Representative changes in left ventricular longitudinal strain (LV-LS). A 60-year-old woman previously
treated with epirubicin demonstrated reduction in global LV-LS at 6 months (–22.4% to –18.9%) concomitant with
a decrease in left ventricular ejection fraction (LVEF). A. Baseline: basal level; B. Baseline: middle level; C. Six month:
basal level; D. Six month: middle level.
Table 2. Reproducibility of peak systolic myocardial strain measured by strain-encoded cardiac
magnetic resonance imaging.
Intra-observer variability
Inter-observer variability
Coefficient of variation
LV, longitudinal
0.968
0.939
–7.20%
LV, circumferential
0.926
0.871
–10.34%
RV, longitudinal
0.931
0.797
–5.94%
RV, circumferential
0.905
0.757
–12.17%
Intra- and inter-observer variability was evaluated using Pearson’s correlation; LV — left ventricle; RV — right ventricle
diography had been used to measure LVEF (61.9%
and 55.9%, respectively).
Role of myocardial strain in the assessment
of chemotherapy-related cardiotoxicity
Myocardial strain measured by echocardiography has been used to detect subclinical myocardial
276
dysfunction in patients with breast cancer receiving
chemotherapy [6–9]. Fallah-Rad et al. [7] reported
that the global LV longitudinal and radial peak
systolic strains decreased at 3 months, followed by
a reduction in LVEF at 6 months in those who developed trastuzumab-related cardiotoxicity. Ho et
al. [9] reported that the global LV longitudinal peak
www.cardiologyjournal.org
Shintaro Nakano et al., CMR for trastuzumab-related cardiotoxicity
systolic strain was lower in asymptomatic breast
cancer patients receiving anthracycline than in
controls, even though the LVEF was broadly comparable between the two groups. Negishi et al. [6]
proposed that changes in global LV longitudinal
peak systolic strain could predict later reduction
in LVEF in patients receiving trastuzumab, when
taken into account alongside established clinical
predictors. We found that CMR-based global LV
peak systolic longitudinal and circumferential
strains were slightly reduced at 3 months, followed
by significant reductions at 6 months, suggesting
that CMR-based myocardial strain studies may be
effective means of detecting asymptomatic early
trastuzumab-related cardiotoxicity. Specifically,
the global LV longitudinal peak systolic strain,
which closely reflects LV dysfunction [31], was the
most reproducible parameter with a relatively low
coefficient of variation, underlining its potential for
the detection of subtle alterations in myocardial
function attributable to trastuzumab. Changes in
LV peak systolic strain were analogous to those
in LVEF measured by CMR, but the predictive
value of strain parameters for later LV dysfunction
could not be determined. Our cohort included only
asymptomatic patients whose LV function did not
markedly decline during the observation period.
The RV free wall circumferential strain had
also decreased by 6 months, but a firm conclusion
about RV function cannot be drawn owing to the
relatively low reproducibility of the test and the
limited number of patients. The importance of RV
function in predicting outcome in heart failure has
been established [32]. Grover et al. [13] reported
that RVEF decreased during anthracycline and/or
trastuzumab therapy, and emphasized the importance of assessment of biventricular function in
cardiotoxicity. Using our method, comprehensive
evaluation of the RV parameters including myocardial strains may further illuminate the mechanisms
of trastuzumab-related cardiotoxicity.
window in some patients; the relative merits
of each technique are controversial and require
further study [30]. Because the measurement of
myocardial strains using echocardiography requires
an adequate acoustic window, we did not compare
strains measured by echocardiography with those
measured by CMR. In our opinion, the limitations
of echocardiography make a strong case for the
use of CMR.
Limitations of the study
Our study has several limitations. First, the
cohort was small, because only those consenting to participate in our study and not any other
potentially conflicting study could be enrolled.
However, the use of CMR-based strain studies in
trastuzumab-related cardiomyopathy is novel and
our findings are sufficient to suggest that CMR
could have a role to play in routine clinical practice. Second, the strain parameters measured by
CMR and echocardiography were not compared
owing to the poor acoustic echocardiographic
5.
Conclusions
We describe the use of CMR-based biventricular myocardial strain studies for the serial
evaluation of cardiac function in patients with
early or locally advanced breast cancer receiving
trastuzumab. Global LV peak systolic longitudinal and circumferential strains had significantly
decreased by 6 months before recovering by
12 months; these changes were analogous to those in
LVEF measured by CMR. RV parameters remained
unchanged except for a marginal decrease in peak
systolic circumferential strains.
Cardiac magnetic resonance imaging-based
assessment of myocardial strain thus appears to
be a useful means of detecting cardiotoxicity in
patients receiving trastuzumab.
Conflict of interest: None declared
References
1.
2.
3.
4.
6.
7.
de Azambuja E, Procter MJ, van Veldhuisen DJ et al. Trastuzumab-associated cardiac events at 8 years of median follow-up in the herceptin adjuvant trial (big 1-01). J Clin Oncol, 2014; 32: 2159–2165.
Tocchetti CG, Ragone G, Coppola C et al. Detection, monitoring,
and management of trastuzumab-induced left ventricular dysfunction: An actual challenge. Eur J Heart Failure, 2012; 14: 130–137.
Chavez-MacGregor M, Zhang N, Buchholz TA et al. Trastuzumab-related cardiotoxicity among older patients with breast cancer.
J Clin Oncol, 2013; 31: 4222–4228.
Seidman A, Hudis C, Pierri MK et al. Cardiac dysfunction in the
trastuzumab clinical trials experience. J Clin Oncol, 2002; 20:
1215–1221.
Otterstad JE, Froeland G, St John Sutton M, Holme I. Accuracy
and reproducibility of biplane two-dimensional echocardiographic
measurements of left ventricular dimensions and function. Eur
Heart J, 1997; 18: 507–513.
Negishi K, Negishi T, Hare JL, Haluska BA, Plana JC, Marwick
TH. Independent and incremental value of deformation indices
for prediction of trastuzumab-induced cardiotoxicity. J Am Society Echocardiography, 2013; 26: 493–498.
Fallah-Rad N, Walker JR, Wassef A et al. The utility of cardiac
biomarkers, tissue velocity and strain imaging, and cardiac magnetic resonance imaging in predicting early left ventricular dysfunction in patients with human epidermal growth factor receptor ii-positive breast cancer treated with adjuvant trastuzumab
therapy. J Am Coll Cardiol, 2011; 57: 2263–2270.
www.cardiologyjournal.org
277
Cardiology Journal 2016, Vol. 23, No. 3
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
278
Hare JL, Brown JK, Leano R, Jenkins C, Woodward N, Marwick TH.
Use of myocardial deformation imaging to detect preclinical myocardial dysfunction before conventional measures in patients undergoing breast cancer treatment with trastuzumab. Am Heart J,
2009; 158: 294–301.
Ho E, Brown A, Barrett P et al. Subclinical anthracycline- and
trastuzumab-induced cardiotoxicity in the long-term follow-up
of asymptomatic breast cancer survivors: A speckle tracking
echocardiographic study. Heart (British Cardiac Society), 2010;
96: 701–707.
Oyama-Manabe N, Sato T et al. The strain-encoded (SENC) MR
imaging for detection of global right ventricular dysfunction in
pulmonary hypertension. Int J Cardiovasc Imaging, 2013; 29:
371–378.
Fallah-Rad N, Lytwyn M, Fang T, Kirkpatrick I, Jassal DS. Delayed contrast enhancement cardiac magnetic resonance imaging
in trastuzumab induced cardiomyopathy. J Cardiovasc Magnetic
Resonance, 2008; 10: 5.
Walker J, Bhullar N, Fallah-Rad N et al. Role of three-dimensional
echocardiography in breast cancer: Comparison with two-dimensional echocardiography, multiple-gated acquisition scans, and cardiac magnetic resonance imaging. J Clin Oncol, 2010; 28: 3429–3436.
Grover S, Leong DP, Chakrabarty A et al. Left and right ventricular effects of anthracycline and trastuzumab chemotherapy:
A prospective study using novel cardiac imaging and biochemical
markers. Int J Cardiol, 2013; 168: 5465–5467.
Korosoglou G, Youssef AA, Bilchick KC et al. Real-time fast
strain-encoded magnetic resonance imaging to evaluate regional
myocardial function at 3.0 tesla: Comparison to conventional
tagging. J Magnetic Resonance Imaging, 2008; 27: 1012–1018.
Teichholz LE, Kreulen T, Herman MV, Gorlin R. Problems in
echocardiographic volume determinations: Echocardiographicangiographic correlations in the presence of absence of asynergy.
Am J Cardiol, 1976; 37: 7–11.
Folland ED, Parisi AF, Moynihan PF, Jones DR, Feldman CL, Tow
DE. Assessment of left ventricular ejection fraction and volumes
by real-time, two-dimensional echocardiography. A comparison
of cineangiographic and radionuclide techniques. Circulation,
1979; 60: 760–766.
Alfakih K, Plein S, Thiele H, Jones T, Ridgway JP, Sivananthan MU.
Normal human left and right ventricular dimensions for mri as
assessed by turbo gradient echo and steady-state free precession imaging sequences. J Magnetic Resonance Imaging, 2003;
17: 323–329.
Kramer CM, Barkhausen J, Flamm SD, Kim RJ, Nagel E. Standardized cardiovascular magnetic resonance imaging (CMR) protocols, society for cardiovascular magnetic resonance: Board
of trustees task force on standardized protocols. J Cardiovasc
Magnetic Resonance, 2008; 10: 35.
Osman NF, Sampath S, Atalar E, Prince JL. Imaging longitudinal
cardiac strain on short-axis images using strain-encoded MRI.
Magnetic Resonance Medicine, 2001; 46: 324–334.
Youssef A, Ibrahim el SH, Korosoglou G, Abraham MR, Weiss RG,
Osman NF. Strain-encoding cardiovascular magnetic resonance
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
for assessment of right-ventricular regional function. J Cardiovasc Magnetic Resonance, 2008; 10: 33.
Cerqueira MD, Weissman NJ, Dilsizian V et al., American Heart
Association Writing Group on Myocardial S, Registration for Cardiac I. Standardized myocardial segmentation and nomenclature
for tomographic imaging of the heart. A statement for healthcare
professionals from the cardiac imaging committee of the council
on clinical cardiology of the american heart association. Circulation, 2002; 105: 539–542.
McMurray JJ, Adamopoulos S, Anker SD et al. Guidelines ESCCfP.
Esc guidelines for the diagnosis and treatment of acute and chronic
heart failure 2012: The task force for the diagnosis and treatment
of acute and chronic heart failure 2012 of the european society of
cardiology. Developed in collaboration with the heart failure association (HFA) of the esc. Eur J Heart Failure, 2012; 14: 803–869.
Lawley C, Wainwright C, Segelov E, Lynch J, Beith J, McCrohon J.
Pilot study evaluating the role of cardiac magnetic resonance
imaging in monitoring adjuvant trastuzumab therapy for breast
cancer. Asia-Pacific J Clin Oncol, 2012; 8: 95–100.
Friedrich MG, Sechtem U, Schulz-Menger J et al., International
Consensus Group on Cardiovascular Magnetic Resonance in M.
Cardiovascular magnetic resonance in myocarditis: A JAAC
white paper. J Am Coll Cardiol, 2009; 53: 1475–1487.
Ewer MS, Lippman SM. Type ii chemotherapy-related cardiac
dysfunction: Time to recognize a new entity. J Clin Oncol, 2005;
23: 2900–2902.
Serrano C, Cortes J, De Mattos-Arruda L et al. Trastuzumabrelated cardiotoxicity in the elderly: A role for cardiovascular risk
factors. Ann Oncol, 2012; 23: 897–902.
Truong J, Yan AT, Cramarossa G, Chan KK. Chemotherapy-induced cardiotoxicity: Detection, prevention, and management.
Canadian J Cardiol, 2014; 30: 869–878.
Strohm O, Schulz-Menger J, Pilz B, Osterziel KJ, Dietz R, Friedrich MG. Measurement of left ventricular dimensions and function in patients with dilated cardiomyopathy. J Magnetic Resonance Imaging, 2001; 13: 367–371.
Gardner BI, Bingham SE, Allen MR, Blatter DD, Anderson JL.
Cardiac magnetic resonance versus transthoracic echocardiography
for the assessment of cardiac volumes and regional function after
myocardial infarction: An intrasubject comparison using simultaneous intrasubject recordings. Cardiovasc Ultrasound, 2009; 7: 38.
Bellenger NG, Burgess MI, Ray SG et al. Comparison of left
ventricular ejection fraction and volumes in heart failure by echocardiography, radionuclide ventriculography and cardiovascular
magnetic resonance; are they interchangeable? Eur Heart J,
2000; 21: 1387–1396.
Geyer H, Caracciolo G, Abe H et al. Assessment of myocardial
mechanics using speckle tracking echocardiography: Fundamentals and clinical applications. J Am Soc Echocardiogr, 2010; 23:
351–369; quiz 453–355.
Ghio S, Gavazzi A, Campana C et al. Independent and additive
prognostic value of right ventricular systolic function and pulmonary artery pressure in patients with chronic heart failure. J Am
Coll Cardiol, 2001; 37: 183–188.
www.cardiologyjournal.org
Shintaro Nakano et al., CMR for trastuzumab-related cardiotoxicity
SUPPLEMENTARY MATERIALS
Detailed definition of data at baseline
Definition of data collection at baseline
Hypertension was defined as current or previous treatment with anti-hypertensive medication.
Diabetes mellitus was defined as current or previous treatment with antidiabetic medication (insulin
or oral hypoglycemic drugs) or a hemoglobin A1c
concentration of ≥ 6.5% [S1]. Dyslipidemia was
defined as current or previous treatment with
anti-dyslipidemic medication, a fasting serum
low-density lipoprotein cholesterol concentration
≥ 160 mg/dL, high-density lipoprotein cholesterol
concentration £ 40 mg/dL, or total cholesterol
concentration ≥ 240 mg/dL [S2].
Normal values of biomarkers
Lower limit of detection levels and upper limit of
normal values were 0.02–0.30 ng/mL for troponin I,
1.0–2.50 ng/mL for myosin light chain, and 4.0–18.40
pg/mL for B-type natriuretic peptide. Values below
the lower limit of detection were treated as zero.
References
S1.
S2.
International expert committee report on the role of the a1c assay
in the diagnosis of diabetes. Diabetes Care, 2009; 32: 1327–1334.
Expert Panel on Detection E, Treatment of High Blood Cholesterol in A. Executive summary of the third report of the national
cholesterol education program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults
(adult treatment panel i ii). JAMA, 2001; 285: 2486–2497.
Supplementary Table S1. Detailed cardiac magnetic resonance protocol.
Study
Cine
Repetition
time [ms]
Echo time
[ms]
Flip angle
[°]
Pixel area
[mm]
Voxel size* Slice thickness Inter-slice gap
[mm]
[mm]
[mm]
2.8
1.42
50
2.97 × 1.69
0.74 × 0.74
8
2
2000
70
90
1.48 × 1.97
0.74 × 0.74
10
–
LGE
3.1
1.49
15
1.48 × 1.90
0.74 × 0.74
3
–
SENC
25
0.88
30
3.0 × 3.0
1.0 × 1.0
10
–
T2 STIR
*Reconstruction voxel size; STIR — short tau inversion recovery; LGE — late-gadolinium enhancement; SENC — strain-encoded imaging
Supplementary Table S2. Echocardiographic parameters at each time point.
Baseline
3 months
P
6 months
P
12 months
P
LV end-diastolic dimension [mm]
43.2 ± 2.9
45.6 ± 1.9
0.069
45.3 ± 3.0
0.084
45.0 ± 2.8
0.088
LV end-systolic dimension [mm]
25.8 ± 1.8
28.1 ± 3.4
0.072
28.1 ± 2.9
0.072
28.5 ± 2.2 0.001*
LV ejection fraction [%]
68.9 ± 6.0 0.037* 66.4 ± 5.9
70.8 ± 5.6
68.7 ± 7.1
0.362
Intra-ventricular wall thickness [mm]
9.5 ± 1.1
9.0 ± 1.3
0.252
9.4 ± 0.7
0.701
9.0 ± 1.3
0.113
0.369
LV posterior wall thickness [mm]
9.3 ± 0.9
9.4 ± 1.3
0.767
9.6 ± 1.7
0.270
8.9 ± 1.3
0.463
Continuous variables are reported as the mean ± standard deviation; *P < 0.05 vs. baseline; LV — left ventricle
www.cardiologyjournal.org
279
Cardiology Journal 2016, Vol. 23, No. 3
Supplementary Table S3. Cardiac magnetic resonance parameters at each time point.
Baseline
3 months
P
6 months
P
12 months
P
LV end-diastolic volume index [mL/m2]
56.8 ± 7.5
59.7 ± 7.0
0.114
58.4 ± 3.7
0.424
59.0 ± 7.8
0.425
LV end-systolic volume index [mL/m2]
24.8 ± 6.7
28.4 ± 7.2 0.024* 30.7 ± 5.4 0.004* 30.5 ± 5.7 0.002*
LV ejection fraction [%]
68.4 ± 6.6
65.4 ± 8.7
0.176
61.7 ± 8.7 0.004* 62.9 ± 7.8 0.017*
LV wall mass index [g/m2]
23.4 ± 4.7
22.9 ± 4.3
0.761
25.2 ± 4.3
Conventional parameters
0.343
23.2 ± 5.1
0.934
2
RV end-diastolic volume index [mL/m ]
68.7 ± 16.3 72.4 ± 15.0 0.256 72.0 ± 12.3 0.443 73.0 ± 14.9 0.204
RV end-systolic volume index [mL/m2]
46.0 ± 13.8 46.8 ± 13.2 0.780 47.2 ± 11.8 0.703 46.5 ± 11.6 0.876
RV ejection fraction [%]
30.7 ± 9.0
31.3 ± 8.5
0.888
31.4 ± 7.3
0.880
31.0 ± 7.4
0.814
Peak systolic strain
Global LV, longitudinal [%]
–21.1 ± 1.7 –20.1 ± 1.5 0.128 –19.5 ± 1.0 0.039* –20.7 ± 1.2 0.622
Global LV, circumferential [%]
–23.4 ± 1.8 –21.9 ± 2.7 0.124 –21.6 ± 2.5 0.036* –21.8 ± 2.3 0.080
RV free wall, longitudinal [%]
–23.3 ± 1.3 –22.6 ± 1.2 0.168 –22.3 ± 1.5 0.096 –22.7 ± 1.7 0.384
RV free wall, circumferential [%]
–20.9 ± 2.4 –21.2 ± 2.4 0.749 –19.1 ± 2.3 0.049* –20.0 ± 1.6 0.276
Continuous variables are reported as the mean ± standard deviation; *P < 0.05 vs. baseline; LV — left ventricle; RV — right ventricle
Supplementary Figure S1. Linear regression plots comparing left-ventricular ejection fraction (LVEF) measured by
cardiac magnetic resonance (CMR) and echocardiography; A. Modified Simpson method; B. Teichholz method;
Pearson’s correlation analysis.
280
www.cardiologyjournal.org